未遵循指令 失败
看看哪些 AI 模型最常遇到 未遵循指令，让你在选择前先发现稳定性风险。 排序方式: 总成本 ↑.
293/293
筛选模型
没有模型匹配当前搜索和筛选条件。
| 排名 | 模型 | 公司 | 未遵循指令 次数 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #269 | GLM 5 Turbo medium | Z.ai | 1 | 0.6 | $0.323 | 14/21 | 23.0s |
| #351 | MiniMax M2.7 medium | Minimax | 5 | 0.5 | $0.326 | 5/23 | 50.3s |
| #248 | MiMo-V2.5-Pro medium | Xiaomi | 2 | 0.6 | $0.332 | 11/23 | 56.7s |
| #341 | Mimo V2 PRO medium | Xiaomi | 1 | 0.5 | $0.333 | 12/21 | 22.2s |
| #82 | Nemotron 3.5 Super VL high | NVIDIA | 2 | 0.8 | $0.349 | 15/23 | 49.9s |
| #36 | DeepSeek V4.1 Flash 1017 low | DeepSeek | 1 | 0.9 | $0.355 | 16/23 | 32.5s |
| #297 | GPT-5.6 Terra none | OpenAI | 1 | 0.6 | $0.377 | 8/23 | 4.11s |
| #298 | DeepSeek V4.1 Flash none | DeepSeek | 2 | 0.6 | $0.391 | 7/23 | 16.6s |
| #157 | Inkling Small high | Thinkingmachines | 2 | 0.7 | $0.398 | 14/23 | 29.1s |
| #213 | MiniMax M3 medium | Minimax | 2 | 0.7 | $0.422 | 11/23 | 77.0s |
| #245 | Beam 1.0 501B A23B low | Reflection | 2 | 0.6 | $0.432 | 11/23 | 4.89s |
| #137 | Nemotron 3.5 Super VL medium | NVIDIA | 1 | 0.8 | $0.437 | 14/23 | 84.9s |
| #99 | Gemini 4 Argon low | 1 | 0.8 | $0.438 | 15/23 | 5.72s | |
| #393 | MiniMax M2.5 medium | Minimax | 3 | 0.4 | $0.440 | 5/23 | 102.3s |
| #167 | Gemini 3.5 Flash Lite medium | 1 | 0.7 | $0.440 | 14/23 | 8.68s |